Daniel-Constantin Mierla writes:

...

$ make modules=modules/<name of module> modules

This should still work, either executed in root folder or in src/. Most of commands should just work as in the past. I tried on Debian and MacOS during the restructuring and worked for me.

OK, works from src, not root.

-- Juha

Daniel-Constantin Mierla writes:

What's the error you get?

No error. Make from root claims that modules is up to date (see below).

-- Juha

/usr/src/orig/kamailio$ ls src/modules/lcr/ doc/ hash.h lcr_mod.h lcr_rpc.h Makefile utils/ hash.c lcr_mod.c lcr_rpc.c makecfg.lst README /usr/src/orig/kamailio$ make modules=modules/lcr modules make: 'modules' is up to date. /usr/src/orig/kamailio$ cd src /usr/src/orig/kamailio/src$ make modules=modules/lcr modules CC (gcc) [M lcr.so] lcr_mod.o CC (gcc) [M lcr.so] lcr_rpc.o CC (gcc) [M lcr.so] hash.o CC (gcc) [L libsrdb1.so.1.0] db_query.o CC (gcc) [L libsrdb1.so.1.0] db_id.o CC (gcc) [L libsrdb1.so.1.0] db.o CC (gcc) [L libsrdb1.so.1.0] db_row.o CC (gcc) [L libsrdb1.so.1.0] db_res.o CC (gcc) [L libsrdb1.so.1.0] db_pool.o CC (gcc) [L libsrdb1.so.1.0] db_ut.o CC (gcc) [L libsrdb1.so.1.0] db_val.o LD (gcc) [L libsrdb1.so.1.0] libsrdb1.so.1.0 LD (gcc) [M lcr.so] lcr.so

Daniel-Constantin Mierla writes:

It seems that it matches 'modules' as filename (dir). I pushed a small change to not enclose the target in quotes when forwarding. Can you try again?

If it still fails, then tell the version of make you have as well as the operating system.

Still the same thing. This in on Debian Jessie. I noticed that I still have this:

/usr/src/orig/kamailio$ ls modules db_mysql/ rtpengine/ sdpops/ tls/ tmx/ websocket/ xmlrpc/

Why does modules dir also exist at root?

-- Juha

/usr/src/orig/kamailio$ eg pull remote: Counting objects: 36, done. remote: Compressing objects: 100% (15/15), done. remote: Total 36 (delta 26), reused 31 (delta 21), pack-reused 0 Unpacking objects: 100% (36/36), done.

From https://github.com/kamailio/kamailio

2b39d70..f27df56 master -> origin/master * [new branch] nsq-max-in-flight -> origin/nsq-max-in-flight Updating 2b39d70..f27df56 Fast-forward .gitignore | 2 +- Makefile | 16 ++++++++++++++-- misc/tools/protoshoot/Makefile | 8 ++++++-- src/Makefile.rules | 3 +++ src/Makefile.utils | 2 +- src/modules/nsq/nsq_mod.c | 12 +++++++++--- 6 files changed, 34 insertions(+), 9 deletions(-) /usr/src/orig/kamailio$ rm src/modules/lcr/*.o /usr/src/orig/kamailio$ rm src/modules/lcr/*.d /usr/src/orig/kamailio$ rm src/modules/lcr/*.so /usr/src/orig/kamailio$ ls src/modules/lcr/ doc/ hash.h lcr_mod.h lcr_rpc.h makecfg.lst README hash.c lcr_mod.c lcr_rpc.c librpath.lst Makefile utils/ /usr/src/orig/kamailio$ make modules=modules/lcr modules make: 'modules' is up to date. /usr/src/orig/kamailio$ ls modules db_mysql/ rtpengine/ sdpops/ tls/ tmx/ websocket/ xmlrpc/ /usr/src/orig/kamailio$ make modules=modules/lcr modules make: 'modules' is up to date. /usr/src/orig/kamailio$ make modules=modules/lcr modules make: 'modules' is up to date. /usr/src/orig/kamailio$ cd src /usr/src/orig/kamailio/src$ make modules=modules/lcr modules CC (gcc) [M lcr.so] lcr_mod.o CC (gcc) [M lcr.so] lcr_rpc.o CC (gcc) [M lcr.so] hash.o CC (gcc) [L libsrdb1.so.1.0] db_query.o CC (gcc) [L libsrdb1.so.1.0] db_id.o CC (gcc) [L libsrdb1.so.1.0] db_row.o CC (gcc) [L libsrdb1.so.1.0] db_val.o CC (gcc) [L libsrdb1.so.1.0] db_res.o CC (gcc) [L libsrdb1.so.1.0] db.o CC (gcc) [L libsrdb1.so.1.0] db_pool.o CC (gcc) [L libsrdb1.so.1.0] db_ut.o LD (gcc) [L libsrdb1.so.1.0] libsrdb1.so.1.0 LD (gcc) [M lcr.so] lcr.so

Mikko Lehto writes:

...

Why does modules dir also exist at root?

Try running "git clean --dry-run". You will propably get some hints of undeleted files.

'git clean -d' deleted modules and lib dirs at root. after that 'make modules=modules/<module> modules' worked from root. wonder why the dirs they were not deleted automatically by 'git pull' when they were removed from root.

-- juha